The use of multiple antennas at a transmitter and/or a receiver of a node in a wireless communication system can significantly boost the capacity and coverage of the wireless communication system. Such Multiple Input Multiple Output (MIMO) systems exploit the spatial dimension of the communication channel to improve performance by for example transmitting several parallel information carrying signals. By adapting the transmission to the current channel conditions, significant additional gains can be achieved. One form of adaptation is to dynamically, from one Transmission Time Interval (TTI) to another, adjust the number of simultaneously transmitted information carrying signals to what the channel can support. This is commonly referred to as (transmission) rank adaptation. Precoding is another form of adaptation where the phases and amplitudes of the aforementioned signals are adjusted to better fit the current channel properties. The signals form a vector-valued signal and the adjustment can be thought of as multiplication by a precoder matrix. A common approach is to select the precoder matrix from a finite and countable set, a so-called codebook. Such codebook based preceding is an integral part of the Long Term Evolution LTE standard and will be supported in MIMO for High Speed Downlink Packet Access (HSDPA) in Wideband Code Division Multiple Access (WCDMA) as well.
Codebook based preceding is a form of channel quantization. A typical approach (c.f. LTE and MIMO HSDPA) is to let the receiver recommend a suitable precoder matrix to the transmitter by signalling the precoder index over a feedback link. The transmitter may choose to directly use the receiver's recommendation, without modification, or to override the receiver's recommendation so signalling the precoder index actually used in the transmission to the receiver may also be needed. To limit signalling overhead, it is generally important to keep the codebook size as small as possible. This however needs to be balanced against the performance impact since with a larger codebook it is possible to better match the current channel conditions.
The design of precoding codebooks and transmission rank adaptation possibilities are a compromise in order to cover a multitude of scenarios. Thus, in certain scenarios, not all combinations of codebook element and transmission rank are beneficial to use. Still, since in practice the adaptation process is non-ideal such ill-suited combinations might be erroneously chosen and thus hurt performance in various ways. The risk for such errors may increase with codebook size and transmission rank possibilities.
The use of adaptation by means of precoding and rank adaptation typically introduces variations in the spatial properties of the transmitted signals. In cellular systems like LTE and WCDMA this can lead to bursty interference which in turn makes link adaptation as well as scheduling more challenging. This issue tends to be more problematic when the number of adaptation combinations is large than when it is small. Letting the receiver select among many different adaptation possibilities means that each possible transmission mode needs to be evaluated in order to find the optimum. This may however involve substantial computational complexity at the receiver and possibly also at the transmitter.
In LTE and WCDMA, a user equipment (UE) selects a recommended precoder and transmission rank and informs the base station, via a feedback channel, about which element in the codebook was selected. Channel Quality Indicators (CQIs) are also fed back conditioned on a certain choice of rank and precoder. The base station may choose to follow the user equipment recommendation or to override it. The problem with the latter is that it increases the CQI errors since the CQIs are computed assuming the use of the recommended precoder and rank. Thus utilizing a larger than necessary codebook and then overriding the user equipment recommendations to limit the transmission freedom, and/or correct erroneous UE selection is not an attractive approach. Using a small codebook and possibly limited rank adaptation with associated reduced signalling overhead is an obvious way of limiting the adaptation possibilities and thus alleviate the above mentioned problems. The problem is that such an approach may create a need for an unnecessarily large number of different codebooks and signalling schemes and that might be highly undesirable from a standard point of view (e.g., complicate performance testing, increase the number of options in the system).